Problem with contact parts

Good afternoon All,

I’m having some trouble with a simulation of two components that are in contact each others.

This are the components (I did a compound of the two):

And this are the F.E. setup that I did:

The surfaces that are in contact are:

I created two specific sufaces to better create the master/slave contact:

I would like to make a static analysis only considering the load gravity; fixig the model on the two lateral extremities.

This is the message that the solver produce after the calculation:

I try many option just to see if I’m making something wrong but everytime the result is the same.

Could you please help me? Because this simple model is a part of a bigger model that I need to calculate (with lot of contacts).

WBR

Marco

Can you share the .pmx file ? This is a generic error only indicating non-convergence (often occurring in underconstrained models involving contact).

How can I reduce the size of the files?

Because at the moment is >10MB

Use File → Close All Results. If that doesn’t help, try making the mesh coarser or using some hosting website where bigger files are accepted (such as Google Drive, Dropbox or WeTransfer, for instance).

ST-studio_prova.pmx (2.1 MB)

Thank you for the suggestion, here the sim file

WBR

Marco

I assume that you made the mesh coarser to share the file, as I suggested (it needs a refinement).

The model isn’t underconstrained because the top part is also fixed (I assume it’s necessary in your setup and not unintentional overconstraint) but hard frictional contact may often fail to converge. You should use linear contact with lower stiffness. One good way to determine it is to use the following approach: Snap-fit contact snagging problem - #17 by JuanP74 - CalculiX (official versions are on www.calculix.de, the official GitHub repository is at https://github.com/Dhondtguido/CalculiX).

This formula gives contact stiffness of 42000 N/mm^3 for the current coarse mesh and that suffices to achieve convergence:

You should also keep in mind the master-slave assignment rules when refining the mesh (master should have a coarser mesh).

1 Like

Yes, I upload a model with a coarse mesh to decrease the size of the files. Very interesting the link shared, I will check it and try to use it different K values.

Thank you very much for the help.

WBR

Marco